Date of Award


Document Type

Restricted Access Thesis

Degree Name

Master of Science in Biology


Department of Biological Sciences

Thesis Advisor

Leonard Reitsma

Committee Member

Kerry Yurewicz

Committee Member

J. Daniel Lambert


The breeding bird communities associated with the southern regions of the North American boreal forest are among the richest and most diverse in North America (Hobson and Bayne 2000). Migratory species in the Northeastern United States comprise between 80% and 90% of the breeding birds, many of which are Neotropical or long distance migrants (Peters et al. 2005).

Birds are commonly the focus of population level studies. Many of these population studies focus upon habitat selection (Block et al. 1987) as it is one of the determining factors of avian distribution and abundance. However, geographic variations in habitat use within a single species may differ from region to region (Graves 2002). To date there is no published research that solely focuses on the habitat selection of the Canada warbler (Wilsonia Canadensis Lambert and Faccio 2005), despite the Northeastern U.S. regional decline in the breeding population. The focus of this thesis is on habitat selection, reproductive success, return rates, and site fidelity of Canada warblers. This study also focuses on the differences of male Canada warbler territories, success rates, return rates, and site fidelity between a relatively mature red maple (Acer rubrum) swamp, and early successional forest directly north of the swamp.

This study was conducted on Canaan Town Forest and Bear Pond Natural Area in central New Hampshire (43.690, -72.0371). From 2003-2006, male territories were delineated using handheld Global Positioning Systems (GPS) between May 27 and July 10. Over 94% of territorial males were banded throughout the study (87 of 92). Pairing and fledging success was documented for each male. Vegetation characteristics were sampled both within male territories and non-territories. These vegetation characteristics were used to determine which habitat features best predict the presence of Canada warblers. Kullback-Leibler information and the Akaike information criterion (AIC) were used to evaluate the vegetation characteristics models, and AICc was used to correct for small sample sizes (Burnham and Anderson 2001).

Shrub density was the best predictor for the presence of Canada warblers. The total number of shrubs, the number of perch sites, canopy height, and the shrub density from 2-2.5m in height was also a strong predictor of Canada warbler presence. The total number of shrubs did not differ between the red maple swamp and the early successional forest. The average territory size was smaller in the red maple swamp and larger in the early successional forest. The reproductive rates on the study plots were high in comparison to other Neotropical migratory warblers that breed throughout the eastern United States. After second year (ASY) males paired at higher rates than second year (SY) birds but there was no difference in the fledging rates. The return rates (50% - 49 of 98) were higher than both the American redstart (Setophaga ruticilla) and the black-throated blue warbler (Dendroica caerulescens) breeding in Hubbard Brook Experimental Forest, but similar to the hooded warbler (Wilsonia citrina) breeding in northwestern Pennsylvania.

The most important vegetative habitat characteristic for Canada warblers is the density of shrubs. The shrub density did not differ between a red maple swamp and an early successional forest created by forest management. Forest management practices create habitat that is ephemeral and may not sustain the high shrub stem densities Canada warblers prefer. The reproductive success also did not differ between the two habitat types. These results suggest that early successional stands create an understory that enables Canada warblers to breed successfully.